According to Appendix Table 1 (Services) of the “Balance of Payments for August 2026 (Preliminary)” released by the Ministry of Finance and the Bank of Japan on October 8, 2026, Japan’s travel account in August recorded receipts of ¥725.4 billion and payments of ¥325.4 billion, for a surplus of ¥399.9 billion. Against the same month a year earlier (receipts ¥701.6 billion, payments ¥339.2 billion, balance ¥362.4 billion), receipts rose +3.4%, payments fell ▲4.1%, and the surplus widened +10.3%. In the same August, international visitor arrivals were 3,098,900 (▲9.6% YoY). Receipts rising while arrivals fell by nearly 10% is the defining feature of the month. Taking balance-of-payments “receipts” as the anchor, this article calculates receipts per visitor (receipts divided by arrivals) for August 2026 and August 2025, organizes the coexistence of fewer visitors and higher receipts into three candidate contributors — exchange rates, market mix and length of stay — and then, as the accommodation-side counterpart, sets the Japan Tourism Agency’s foreign guest-night figures alongside our estimated settled ADR for five prefectures.

August Travel Balance — Receipts ¥725.4bn, Payments ¥325.4bn, Surplus ¥399.9bn

Let us first confirm the figures from the primary source as they stand. In Appendix Table 1 “Services” of the Ministry of Finance / Bank of Japan release, the travel account for August 2026 showed receipts of ¥725.4 billion against payments of ¥325.4 billion, for a surplus of ¥399.9 billion. A year earlier, receipts were ¥701.6 billion, payments ¥339.2 billion and the balance ¥362.4 billion, so receipts increased by ¥23.8 billion (+3.4%) and payments decreased by ¥13.8 billion (▲4.1%). With both moving in the same direction, the surplus widened by ¥37.5 billion (+10.3%). For the services account as a whole, receipts were ¥3.4721 trillion and payments ¥3.5879 trillion, keeping the deficit to ¥115.8 billion, narrower than the ¥266.7 billion of a year earlier.

Source: Compiled by HotelBank Editorial Team from Ministry of Finance / Bank of Japan, “Balance of Payments for August 2026 (Preliminary),” Appendix Table 1 Services (released October 8, 2026). Prior-year figures are from the prior-year column of the same release

Source: Compiled by HotelBank Editorial Team from Ministry of Finance / Bank of Japan, “Balance of Payments for August 2026 (Preliminary),” Appendix Table 1

The reference section of the same release, citing JNTO, lists August international visitor arrivals of 3,098,900 (▲9.6% YoY) and Japanese departures of 1,588,600 (▲3.6%), and records the dollar–yen rate at a monthly average of ¥158.74/USD (¥147.67 a year earlier, a 7.5% depreciation of the yen) and the euro–yen rate at ¥184.13/EUR (¥171.74 a year earlier, a 7.2% depreciation). The combination of rising receipts in a month when arrivals fell by nearly 10% only comes into focus once these two numbers — the exchange rate and arrivals — are placed side by side.

As for the year-to-date trend, Nikkei reported on the same day that the travel surplus for January–August was ¥4.4540 trillion, down 0.6% (¥28.1 billion) year on year. Because each month’s preliminary figures are subsequently revised in the second preliminary release and the annual revision, this article does not simply sum the monthly preliminaries but limits itself to citing this reported figure with its source. The picture of a single month, August, showing double-digit growth while the year to date remains roughly flat is the premise for reading the August numbers.

Receipts per Visitor ¥234,100 — Up +14.4% from ¥204,600 a Year Earlier

So what happens when receipts are divided by the number of visitors? The formula is simple: travel receipts (¥100 million) × 100 million ÷ visitor arrivals (persons). For August 2026, ¥725.4 billion ÷ 3,098,900 = approximately ¥234,100; for August 2025, ¥701.6 billion ÷ 3,428,406 = approximately ¥204,600, making receipts per visitor +14.4% year on year. The numerator includes not only accommodation but also shopping, food and drink and domestic transport, and excludes international airfares (see the note above). The denominator, JNTO visitor arrivals, includes business travelers, students and other entrants as well as tourists, so this value should be read not as “spending per tourist” but as a reference value of “receipts allocated per visitor.”

Source: Compiled by HotelBank Editorial Team from Ministry of Finance / Bank of Japan, “Balance of Payments for August 2026 (Preliminary),” and Japan National Tourism Organization (JNTO), “Visitor Arrivals (August 2026 estimates).” USD conversion is a convenience conversion at the monthly average rate stated in the release

The key point in this calculation is how much of the +14.4% in yen terms can be explained by the exchange rate. Converting back at the monthly average rates stated in the release, receipts per visitor rose +6.4%, from $1,385.8 to $1,474.6, while total receipts fell ▲3.8%, from $4.75 billion to $4.57 billion. In other words, in foreign-currency terms, total receipts did not fall as much as arrivals, and on a per-visitor basis they increased even in foreign-currency terms. Multiplying arrivals ▲9.6%, per-visitor foreign-currency spending +6.4% and yen depreciation +7.5% gives 0.904 × 1.064 × 1.075 ≈ 1.034, matching exactly the +3.4% rise in yen-denominated receipts. This decomposition is a convenience that assumes settlement in dollars, but it serves as a guide for separating the reason receipts rose in a month of falling arrivals into two parts: the exchange rate and per-visitor spending in foreign-currency terms.

Source: Compiled by HotelBank Editorial Team from Ministry of Finance / Bank of Japan, “Balance of Payments for August 2026 (Preliminary),” and JNTO, “Visitor Arrivals (August 2026 estimates).” The USD conversion is a convenience decomposition

Why Receipts Rose as Arrivals Fell — Three Candidate Contributors: Exchange Rates, Market Mix and Length of Stay

The statistics cannot pin down a single cause for receipts rising while arrivals fell. Without asserting causation, we lay out three candidate contributors. The first is the exchange rate discussed in the previous section: even if foreign-currency spending were unchanged, yen-denominated receipts would be inflated by 7.5%. The second is a shift in market mix. According to JNTO’s August estimates, China fell sharply to 418,000 (▲59.0% YoY) from 1,018,747 a year earlier, while Korea reached 850,500 (+28.7%), a record for August; Taiwan 666,000 (+7.3%), Hong Kong 247,300 (+9.4%) and the United States 197,400 (+1.5%) also grew, and Italy 47,300 (+15.1%) and Spain 36,500 (+14.1%) set single-month records. Arrivals excluding China rose +11.3%, from 2,409,659 to 2,680,900, so the “decline in arrivals” was effectively a single market — China — pulling down the total.

Source: Compiled by HotelBank Editorial Team from Japan National Tourism Organization (JNTO), “Visitor Arrivals (August 2026 estimates)” (September 16, 2026). 2025 figures are final; 2026 figures are estimates

How a shift in market mix affects receipts depends on each market’s per-capita spending and length of stay. The Japan Tourism Agency’s “Inbound Consumption Survey” for July–September 2026 (first preliminary, released October 21) reported international visitor spending of ¥2.2076 trillion (+3.2% YoY), per-capita travel spending of ¥233,677 (+6.1%) and an average stay of 10.3 nights. Spending by nationality/region ranked Taiwan ¥381.8 billion (+26.2%), China ¥358.3 billion (▲39.8%), Korea ¥293.8 billion (+45.8%), the United States ¥240.9 billion and Hong Kong ¥155.5 billion (+37.1%). That China’s spending fell only ▲39.8% while its arrivals fell ▲56.7% (January–August cumulative, JNTO) suggests that the travelers who remained stay longer and spend more per head; China’s per-capita travel spending was ¥294,560 (+22.1%), above the overall average.

The third candidate is length of stay. Dividing foreign guest nights by nationality from the Japan Tourism Agency’s “Overnight Travel Statistics Survey” for August 2026 (first preliminary; facilities with 20 or more rooms) by JNTO arrivals for the same month gives guest nights per visitor of 1.9 for Korea, 2.5 for Hong Kong, 3.1 for Taiwan, 4.1 for China, 5.0 for the United States, 7.3 for France, 8.1 for Italy and 8.5 for Spain. In a month when short-staying Korea drove arrival growth and long-staying Europe set records, it naturally follows that changes in arrivals do not map directly onto changes in guest nights or spending.

Source: Compiled by HotelBank Editorial Team from JNTO, “Visitor Arrivals (August 2026 estimates),” and Japan Tourism Agency, “Overnight Travel Statistics Survey,” August 2026 first preliminary, Table 2 (facilities with 20+ rooms, including unknown nationality). Nights per visitor is a reference value of guest nights ÷ arrivals; note that it does not account for gaps between the month of entry and the month of stay, or for facilities with fewer than 20 rooms

Putting this together, the rise in August receipts can be read as the overlap of two explanations: “the weak yen inflated yen-denominated figures” and “the shortfall from China was filled by long-staying Western and Australian travelers, high-spending remaining Chinese travelers, and large volumes from Korea and Taiwan.” Which contribution is larger cannot be determined without data on settlement currencies and per-market spending, but at least in this month, reading the decline in arrivals as an equivalent decline in receipts does not hold.

The Lodging-Side Counterpart — Foreign Guest Nights 12.64 Million (▲10.3%) and Estimated Settled ADR in 5 Prefectures

How does the rise in receipts show up on the ground in accommodation? According to the Japan Tourism Agency’s “Overnight Travel Statistics Survey” for August 2026 (first preliminary, released September 30), total guest nights in August were 62.68 million (▲6.7% YoY), of which Japanese guests accounted for 50.04 million (▲5.8%) and foreign guests 12.64 million (▲10.3%). Overall room occupancy was 64.4% (▲1.3pt YoY); ryokan were 47.4% (+1.7pt), above the prior year, followed by business hotels at 75.7% (▲0.7pt), city hotels at 72.3% (▲3.0pt) and resort hotels at 66.9% (▲0.6pt). The Japan Tourism Agency notes that from the January 2026 survey it changed the stratification criterion for facilities from number of employees to number of rooms, and that YoY comparisons may include the effect of this revision. Foreign guest nights ▲10.3% moved almost in step with arrivals ▲9.6%, meaning the decline in arrivals translated directly into fewer guest nights on the lodging side.

What about rates? Comparing MetroEngines Research’s estimated settled ADR (historical basis for finalized past months, median across all categories) year on year for Tokyo, Osaka, Kyoto, Fukuoka and Hokkaido, August showed Tokyo ▲9.2% (¥13,000→¥11,800, N=1,120/1,140 properties), Osaka ▲32.7% (¥13,500→¥9,100, N=644/653), Kyoto ▲8.4% (¥14,300→¥13,100, N=590/605), Fukuoka ▲2.1% (¥11,500→¥11,300, N=460/476) and Hokkaido ▲5.1% (¥12,900→¥12,200, N=918/960). Osaka’s outsized decline is because August 2025 fell during Expo 2025 Osaka, Kansai, leaving a high prior-year base; it is most naturally read as normalization to post-Expo levels. In September, by contrast, four regions turned positive — Tokyo +1.9%, Kyoto +1.1%, Fukuoka +9.8% and Hokkaido +4.5% — and the fact that the summer decline did not carry into autumn is consistent with the resilience on the receipts side.

Source: Compiled by HotelBank Editorial Team from MetroEngines Research. Estimated settled ADR is on a historical basis for finalized past months, median across all categories (tax-excluded equivalent). September YoY compares estimated settled ADR for September 2025 and September 2026

Source: Compiled by HotelBank Editorial Team from MetroEngines Research (estimated settled ADR, historical basis, N=460–1,140 properties per prefecture)

Balance-of-payments receipts at +3.4%, guest nights at ▲10.3% and room rates negative in key regions — at first glance, these three numbers do not fit together. However, receipts include shopping, food and drink and domestic transport, and accommodation is only one part. In the Japan Tourism Agency’s July–September survey, the breakdown of per-capita spending was ¥82,523 for accommodation, ¥64,367 for shopping and ¥53,600 for food and drink, with accommodation just over one-third of the total. If the increase in receipts is carried by non-accommodation categories, by lodging types other than hotels, or by the conversion effect of the weak yen, there is no contradiction in hotel room rates not directly reflecting it in a given month. For the lodging side, this can be reframed as meaning there is still room to capture the increase in receipts in room rates.

The Payments Side — Japanese Departures ▲3.6% and Room for a Return to Domestic Stays

The decline on the payments side also contributed to the wider surplus. August travel payments were ¥325.4 billion (▲4.1%) and Japanese departures were 1,588,600 (▲3.6%), so payments per departing traveler were approximately ¥204,800, almost flat (▲0.5%) from approximately ¥205,800 a year earlier. That per-capita outbound travel spending has not risen in yen terms amid a weakening yen suggests that travelers may increasingly be adjusting length of stay or destinations to stay within budget. Departures for January–August totaled 9,724,500 (+2.8%), above the prior year, but August alone was down 3.6%, leaving room for part of summer-holiday outbound demand to shift to domestic travel.

The natural recipient of that shift is domestic accommodation. In the Japan Tourism Agency’s August first preliminary, Japanese guest nights were 50.04 million (▲5.8%), below the prior year, so Japanese travelers have not yet filled the gap left by foreign guests. However, if travelers with budgets of over ¥200,000 per overseas trip redirect that spending domestically, it aligns well with high-rate resort and ryokan categories. The fact that ryokan were the only category whose room occupancy exceeded the prior year in August, at 47.4% (+1.7pt), can be read as evidence that this return has begun in some segments.

Implications for Operators — Budget on Receipts per Visitor, Not Arrivals

What the August numbers show operators is that setting inbound revenue budgets solely on the YoY change in visitor arrivals can get the direction wrong in some months. With receipts +3.4% and receipts per visitor +14.4% in a month when arrivals fell 9.6%, it is worth holding two separate budget variables: “arrivals” and “per visitor.” Arrivals can be tracked through JNTO’s monthly estimates (released mid-following month), and per-visitor values through balance-of-payments travel receipts (released early in the second following month) divided by arrivals — both updated monthly using public statistics alone.

The shift in market mix also has implications for room-rate design. August, when Korea (1.9 nights per visitor) drove arrivals and Europe (7–8 nights) set records, was a month in which both short-stay turnover and long-stay multi-night pricing mattered at the same time. Even as Chinese arrivals fell sharply, per-capita spending among those who remained was high at ¥294,560 (+22.1%), so there is demand that can be captured on the rate side while waiting for volumes to recover. On exchange rates, while the weak yen lifts receipts per visitor in yen terms, Japanese accommodation continues to look inexpensive in foreign-currency terms, making a review of price bands with dollar- and euro-denominated perceived prices in mind an opportunity. We examine how rates compare by currency in Weak Yen & Inbound 2026 (currency-based ADR index).

From an investor’s perspective, balance-of-payments travel receipts act as a proxy for total demand including non-accommodation spending, and signal direction two months earlier than prefecture-level guest nights or estimated settled ADR. That September estimated settled ADR turned positive in Tokyo, Kyoto, Fukuoka and Hokkaido following August receipts of +3.4% is worth tracking as a sign that resilient receipts are beginning to feed through into rates. See also our July 2026 volume × rate four-quadrant analysis, which classifies the relationship between guest nights and rates across 47 prefectures, and our Inbound 2026 forecast progress tracker for the annual outlook on visitor arrivals.

Summary

In August 2026, the travel account recorded receipts of ¥725.4 billion, payments of ¥325.4 billion and a balance of ¥399.9 billion — receipts +3.4%, payments ▲4.1% and balance +10.3% year on year. Behind the rise in receipts in a month when arrivals fell 9.6% stand three candidate contributors: the exchange rate (a 7.5% weaker yen), market mix (China ▲59.0%, Korea +28.7%, single-month records for Italy and Spain) and length of stay (Korea 1.9 nights, Europe 7–8 nights). Receipts per visitor rose to ¥234,100 (+14.4%), and +6.4% even in dollar terms. On the lodging side, foreign guest nights were ▲10.3% and August estimated settled ADR was negative in all five prefectures, but four regions returned to positive territory in September. The year-to-date surplus is roughly flat, down 0.6% year on year according to press reports, and the idea of capturing value on a per-visitor basis rather than waiting for arrivals to recover should be the backbone of budgets from autumn onward.

Frequently Asked Questions

Q. What is included in travel “receipts” in the balance of payments?
The total spent by non-residents visiting Japan during their stay on accommodation, food and drink, shopping, domestic transport, entertainment and so on. International airfares are recorded separately under “Transport (passenger)” and are not included in travel.

Q. How does this article’s receipts per visitor differ from the Japan Tourism Agency’s per-capita travel spending?
This article’s value is a reference figure obtained by simply dividing balance-of-payments travel receipts by JNTO visitor arrivals (which include business travelers, students and others). The Japan Tourism Agency’s “Inbound Consumption Survey” is an estimate based on interviews at airports and other points, with a different population and aggregation method. The figures will not match even for the same period.

Q. Why does the prior-year travel balance differ from the preliminary figure published a year ago?
Balance-of-payments statistics are revised retroactively in the second preliminary release and the annual revision. This article uses the prior-year column of the release published by the Ministry of Finance on October 8, 2026, which may not match the preliminary figure as of September 2025.

Q. Which months are used to calculate the YoY change in estimated settled ADR?
We compare August 2025 with August 2026, and September 2025 with September 2026, in each case using estimated settled ADR (median across all categories) on a historical basis for finalized past months. These are tax-excluded-equivalent estimated settled rates, not tax-included listed prices.

Related Articles